Abstract Based on such criteria as safety and mission success, programmatic risk, affordability, and extensibility/flexibility, the National Aeronautics and Space Administration (NASA) has chosen fission surface power (FSP) as the primary energy source for building a sustained human presence on the Moon, exploring Mars, and extremely long-duration space missions. The current benchmark FSP system has a mission life of at least 8 years during which time there is no opportunity for repair, sensor calibrations, or periodic maintenance tasks that are normally performed on terrestrial-based nuclear power plants during scheduled outages. Current technology relies heavily on real-time human interaction, monitoring and control. However; due to the long communication times between the Earth and Moon, or Mars, real-time human control is not possible, resulting in a critical need to develop autonomous health monitoring technology for FSP systems.This paper describes the design and development of an autonomous health monitoring system that will (1) provide on-line calibration monitoring, (2) reduce uncertainties in sensor measurements, and (3) provide sensor validation and fault detection capabilities for the control systems of various FSP subsystems. The health monitoring system design integrates a number of signal processing algorithms and techniques such as cross-calibration, empirical modeling using neural networks, and physical modeling under a modular signal processing platform that will enable robust sensor and system monitoring without the need for human interaction. Prototypes of the health monitoring system have been tested and validated on data acquired from preliminary subsystem testing of NASA's FSP Technology Demonstration Unit (TDU) as well as simulated laboratory data. Results from this testing have demonstrated the utility and benefits that such autonomous health monitoring systems can provide to FSP subsystems and other potential applications within NASA such as launch vehicle systems, other nuclear power systems, and ground operations.
Read full abstract